The formation of barium giants via mass accretion in binary systems

TL;DR

This study models the formation of barium giants through mass transfer from AGB companions, fitting observed star properties with simulated accretion scenarios to understand their surface compositions and evolutionary states.

Contribution

It introduces detailed models of mass accretion from AGB stars onto low-mass companions, matching observed barium star properties and exploring accretion efficiencies and orbital correlations.

Findings

Most barium giants are best fit with accretion from 2.5 or 3 M$_ ext{sun}$ AGB stars.

Accretion masses are consistent with hydrodynamical wind transfer simulations.

No strong correlation between orbital period and accreted mass was found.

Abstract

We examine the composition of barium stars in the context of mass transfer from an asymptotic giant branch (AGB) companion. We accrete between 0.01 and 0.5 M$_\odot$ of AGB ejecta on to low mass companions of [Fe/H] = -0.25 at the ages expected for the end of the lives of AGB stars of 2.5, 3 and 4M$_\odot$. In each case, we form a star of 2.5 M$_\odot$ which is thought to be a typical barium star mass. We discuss the extent of dilution of accreted material as the star evolves, and describe the impact on the surface abundances. For accretion from a 2.5\ms\ primary, if the secondary's initial mass is 2.45 M$_\odot$ or more, accretion takes place when the secondary is undergoing core helium burning. Using data from the sample of De Castro et al., we attempt to fit the observed properties of 74 barium giants using the models we have computed. We find that all but six of these objects are best fit using ejecta from 2.5 M$_\odot$ (32 objects) or 3 M$_\odot$ (36 objects) AGB stars. Higher accretion masses are typically required when accreting from a lower mass companion. We find accretion masses that are broadly consistent with recent hydrodynamical simulations of wind mass transfer, though the accretion efficiency is toward the upper limit found in these simulations. For the 18 stars with reported orbital periods, we find no strong correlations between period and accretion mass.